Merged Puzzling results from CERN

[Walter Wayne]

Not yet, but apparently they are rerunning the experiments with short bursts of neutrinos in order to make the statistics at little easier. The short bursts are intended to allow them to trace neutrinos back to a particular event more reliably.

http://www.bbc.co.uk/news/science-environment-15471118

 

[Soapy Sam]

Twice.Ending back where it started. But for a time, all was topsy turvey.

Yes- some way of tagging individual neutrinos would be damn useful.

Pixie paint, perhaps?

 

[ben m]

How does it know to pick the shortest path?

...

Or it goes every way and we just measure one.

The last of those is pretty close to the truth. When you're dealing with a wave phenomenon, the "shortest path" happens to have the special property that neighboring paths don't destructively interfere with it. All of the non-shortest-path options get destroyed by such an interference.

This was applied to light, at least heuristically, in the Huygens-Fresnel Principle, and applied to quantum fields by Feynman in his "path integral formulation".

 

[Soapy Sam]

I have driven through the Sahara . In part, I followed the course of earlier drivers, who had created an "easiest" route-one I did not need to think about.
In part, I probed ahead of myself by examining photons that had been out ahead of me. (ie I looked through the windshield).

But the latter option is (I presume) unavailable to photons themselves, They can't sniff out the easiest route. They are constrained, either by some aspect of the way they propagate or by some aspect of spacetime, (assuming these are not aspects of the same thing) to follow a given route.
Or they do indeed follow all routes and somehow "come together" at the end in a way that looks as if they followed the shortest route.

A train goes around a curve on the track, not because the rails are curved, but becausethe combined forces acting on it make staying on the rails the only place it can be at any instant.
Crank the speed up enough and that situation will change.

To say light follows curves in space does not (to me) offer an explanation. It merely restates what happens.

If we could make it go fast enough, would light go off the rails?
How about neutrinos?

 

[sol invictus]

If we could make it go fast enough, would light go off the rails?
How about neutrinos?

There's nothing teleological about following the shortest path, so long as the "decision" is local. That's the case for light - it follows what's locally a straight line. Over long distances the path can look curved, because the spacetime is, but it's really the analog of a straight line.

If you find that confusing, just think about an object sliding on smooth, flat ice (or drifting through a vacuum in outer space). How does it "know" to keep going straight? Why does it always take the shortest path between any two points along its path? It's exactly the same for light, it just goes straight (unless you mess with it by putting something in its way, of course).

Light cannot "go off the rails" because it always moves at the same speed. You can't make it go faster or slower. Neutrinos on the other hand have a mass, and they do indeed follow different spacetime paths depending on their velocity. However as their velocity approaches c, as is the case for neutrinos produced at CERN, their path converges to that of light.

 

[Michael Mozina]

There's nothing teleological about following the shortest path, so long as the "decision" is local. That's the case for light - it follows what's locally a straight line. Over long distances the path can look curved, because the spacetime is, but it's really the analog of a straight line.

If you find that confusing, just think about an object sliding on smooth, flat ice (or drifting through a vacuum in outer space). How does it "know" to keep going straight? Why does it always take the shortest path between any two points along its path? It's exactly the same for light, it just goes straight (unless you mess with it by putting something in its way, of course).

Light cannot "go off the rails" because it always moves at the same speed. You can't make it go faster or slower. Neutrinos on the other hand have a mass, and they do indeed follow different spacetime paths depending on their velocity. However as their velocity approaches c, as is the case for neutrinos produced at CERN, their path converges to that of light.

I haven't been following this conversation, but I am curious if you have seen a "solution" to the problem that you're happy with yet?

 

[sol invictus]

I haven't been following this conversation, but I am curious if you have seen a "solution" to the problem that you're happy with yet?

There's a defect in the experimental design - the pulses are 10,000 ns long, and they're trying to measure timing to an accuracy of 10 ns - that seems quite plausibly the source of the error. We discussed it at some length earlier in the thread - the details have to do with the way the data was collected and the statistics involved in the analysis.

While I don't have a way to demonstrate that that is in fact the problem (because I don't have access to the data or sufficient information about some details of the experiment), I'm not going to take the result at all seriously until it gets addressed. I think they are re-doing the experiment with this objection in mind, so we'll see what happens.

 

[Michael Mozina]

There's a defect in the experimental design - the pulses are 10,000 ns long, and they're trying to measure timing to an accuracy of 10 ns - that seems quite plausibly the source of the error. We discussed it at some length earlier in the thread - the details have to do with the way the data was collected and the statistics involved in the analysis.

While I don't have a way to demonstrate that that is in fact the problem (because I don't have access to the data or sufficient information about some details of the experiment), I'm not going to take the result at all seriously until it gets addressed. I think they are re-doing the experiment with this objection in mind, so we'll see what happens.

Cool. Thanks for the update.

 

[Farsight]

I have driven through the Sahara...

Light curves in much the same way as your vehicle veers left when the sand is softer on the left than the right.

To say light follows curves in space does not (to me) offer an explanation. It merely restates what happens.

Yes, it's the effect not the cause. Have a read of Einstein's 1920 Leyden Address and see the bit that says:

"According to this theory the metrical qualities of the continuum of space-time differ in the environment of different points of space-time, and are partly conditioned by the matter existing outside of the territory under consideration. This space-time variability of the reciprocal relations of the standards of space and time, or, perhaps, the recognition of the fact that “empty space” in its physical relation is neither homogeneous nor isotropic..."

Space is rather like that sand. Here's a web page that gets it across quite well: http://www.astro.ucla.edu/~wright/deflection-delay.html . Note the bit that says "In a very real sense, the delay experienced by light passing a massive object is responsible for the deflection of the light". So light travels at variable speed, hence the neutrinos aren't such a big deal. They're as amazing as light travelling faster than light, which is ubiquitous.

 

[sol invictus]

So light travels at variable speed, hence the neutrinos aren't such a big deal. They're as amazing as light travelling faster than light, which is ubiquitous.

Nonsense.

Neutrinos traveling at variable speed is not a big deal (they have mass, for one thing, so their speed is obviously variable). Neutrinos traveling faster than light would in a vacuum if it followed the same path (for instance, if there was a thin pipe full of vacuum connecting CERN and Gran Sasso and you beam some light down it) is a very big deal.

 

[Soapy Sam]

There's nothing teleological about following the shortest path, so long as the "decision" is local. That's the case for light - it follows what's locally a straight line. Over long distances the path can look curved, because the spacetime is, but it's really the analog of a straight line.

How local is "local"?
One of the (very many) things that baffles me is that in QM it seems to be considered reasonable to say a photon follows every possible path to get from point 1 to point 2. So although point 2 is a gazillionth of an inch (1Gz") due north of point 1, the photon actually goes via the south pole and Messier 31, but those paths sort of average out probability-wise, so it looks like the photon really did go the obvious route, 1Gz" due north.
Yet when we talk of light going from M31 to Palomar, we assume it follows a geodesic- either a straight line or a curved one, depending on mass distribution along the route . But the space between M31 and Palomar can be divided into (a very large number of) 1Gz" steps, which presumably means that photon has one helluva lot of routes to explore and a lot of local "decisions" to make.
It surely cannot be that some sort of information processing / decision making is involved. (This may be argument from incredulity, but I mean geez...) so it seems like the photon "just goes where it must".

But if we look at the damn thing through a double slit it may take a different route (for the final inch) than it would if we only look through one slit. Which looks to me like the photon "made a decision" based not on spacetime geometry, but on how we set up the equipment in the telescope. Now did it make that decision locally? Did it choose which slit to go through at the fraction of a second it reached the detector? Or when it left Andromeda?

 

[Dan O.]

It only seams that way after the fact. From the photo's frame of reference, there is no distance between M31 and Palomar or any other possible destination so there is no decision to be made in going, there is only probability of where it will end up.

 

[Michael Mozina]

There's a defect in the experimental design - the pulses are 10,000 ns long, and they're trying to measure timing to an accuracy of 10 ns - that seems quite plausibly the source of the error. We discussed it at some length earlier in the thread - the details have to do with the way the data was collected and the statistics involved in the analysis.

While I don't have a way to demonstrate that that is in fact the problem (because I don't have access to the data or sufficient information about some details of the experiment), I'm not going to take the result at all seriously until it gets addressed. I think they are re-doing the experiment with this objection in mind, so we'll see what happens.

http://www.bbc.co.uk/news/science-environment-15471118

You're right about them rerunning the experiments by the way. I look forward to their new results.

 

[sol invictus]

How local is "local"?

As local as you can get.

One of the (very many) things that baffles me is that in QM it seems to be considered reasonable to say a photon follows every possible path to get from point 1 to point 2.

Forget quantum mechanics. You're question was about classical physics - electromagnetic waves, i.e. light - or at least that's what I assumed.

 

[Soapy Sam]

It was, Sol, but I keep reading books that say forget classical mechanics, light follows quantum rules. If I'm understanding this POV at all, it's saying that at each step on that long journey, the photon "decides" where it goes next in a quantum fashion. The averaged result over the whole path looks like relativity at work.
I keep thinking something is being missed here- certainly by me, but just possibly by everyone else as well.

 

[Vorpal]

It was, Sol, but I keep reading books that say forget classical mechanics, light follows quantum rules. If I'm understanding this POV at all, it's saying that at each step on that long journey, the photon "decides" where it goes next in a quantum fashion. The averaged result over the whole path looks like relativity at work.

A lot of the mysteriousness is alleviated if one realizes that the wavefunction of a single photon would be described exactly by Maxwell's equations, in the form Ψ = E+iB and normalizable. A lot of hairy quantum stuff comes in when one considers interactions with matter, but it is still unwise to forget the classical theory.

 

[chuck4842]

There's a disturbingly teleological aspect to the concept of a shortest route.
To say an object follows a straight line (or any shortest course through a space) we need to know where it started and where it is going.
Imagine I'm walking across a flat plane- a field, say.
I start at point 1 and walk to point 2 where I change direction and walk to point 3.
An observer can say I took the shortest route from 1 to 2 and from 2 to 3- but not from 1 to 3, because I changed direction at 2.
As a conscious entity I can say that I set out to go to 3, so the course I took was not, at any point, the shortest possible - but there MUST always exist a point 2 - and as any 2 points are linkable by a straight line, I am constrained to follow that line whether I want to or not, because if I changed course every inch between points 1 and 2, then there must exist a new point 2, closer to point 1, which I DID reach by the shortest distance possible. In short, no matter what convolutions my course follows, it is a linked chain of straight lines, of possibly very short length- each of them a shortest possible course . On a quantum scale, such line segments may approach the Plank length, yet still, between each two points, there exists a "straight line".

Is this all that is meant by saying that light follows straight lines in spacetime?

Or is this too incoherent to make sense?

I think you're getting close to what happened. Has anyone posted the published report showing the expected path diagram?

http://arxiv.org/ftp/arxiv/papers/1109/1109.4897.pdf

 

[sol invictus]

It was, Sol, but I keep reading books that say forget classical mechanics, light follows quantum rules. If I'm understanding this POV at all, it's saying that at each step on that long journey, the photon "decides" where it goes next in a quantum fashion. The averaged result over the whole path looks like relativity at work.
I keep thinking something is being missed here- certainly by me, but just possibly by everyone else as well.

Hmm. Which books are telling you that, and what exactly do they say? Can you give a quote?

As Vorpal says, the propagation of light through a vacuum is very well described by classical physics.

I think you're getting close to what happened. Has anyone posted the published report showing the expected path diagram?

http://arxiv.org/ftp/arxiv/papers/1109/1109.4897.pdf

So are you going to enlighten us? What happened, according to you?

 

[Soapy Sam]

Hmm. Which books are telling you that, and what exactly do they say? Can you give a quote?

One such was certainly "The Fabric of Reality" by David Deutsch. I've just been looking in my copy for the relevant passage, but no luck yet. I expressed myself poorly in the previous post- I don't mean anyone was saying the classical explanation is invalid, but that any classical explanation should also be explicable in quantum terms. I think most popularisations involving double slit experiments and explanations thereof stress that the quantum view of how photons move is real, despite it's apparent incompatibility with common sense. So I must suppose this applies as much to light crossing between galaxies as light moving through an experimental setup in a lab. It's also very likely I simply misunderstood what was meant.

 

[Vorpal]

I expressed myself poorly in the previous post- I don't mean anyone was saying the classical explanation is invalid, but that any classical explanation should also be explicable in quantum terms. I think most popularisations involving double slit experiments and explanations thereof stress that the quantum view of how photons move is real, despite it's apparent incompatibility with common sense.

If you're just having light in a vacuum, essentially all of the quantum nonconformity to common sense is the fact that the wave is that of probability describing particle. The all-paths summation for some weighting is already how classical waves behave. For light, this was noticed very early on in approximate way Huygens, as ben m said, and more precisely as a certain restatement of the Kirchhoff diffraction integral of classical EM.

This applies just as well to plain-vanilla, nonrelativistic QM. In the eikonal approximation of wave propagation used as an intermediary in deriving geometric (ray, short-wavelength) optics from wave optics, one only has to know about the de Broglie relations to get the standard Schrödinger equation, which also has an equivalent path-integral formulation. All the 'magic' is really in the idea that waves and particles have anything to do with each other (specifically here, de Broglie), not the path-summing. And though that de Broglie connection can be motivated to a large extent by the near-identical mathematical behavior of geometric optics and classical Hamiltonian mechanics, it's still pretty crazy from a common-sense point of view.